PROJECT SUMMARY/ABSTRACT Neuroblastoma (NB) is the most common extracranial solid tumor of childhood. While intensive induction chemotherapy and aggressive surgery have improved remission rates in young patients, results have been less impressive in older patients and in the 40% of patients with chemoresistant NB. Metastatic NB in children > 18 months of age at diagnosis carries a long-term relapse-free survival of only ~20%. These disappointing results are compelling reasons for pursuing novel therapeutic approaches. One alternative approach has been immunotherapy with murine 3F8, an IgG3 monoclonal antibody to the GD2 glycolipid antigen ubiquitously present on NB cells. Its major disadvantage, however, is the development of neutralizing antibodies, which can limit its effectiveness. 3F8 administration can confer a survival benefit to NB patients in remission, but it is not effective for patients with resistant soft tissue NB or patients with progressive disease. This may be due in part to its dependence on antibody-dependent cell-mediated cytotoxicity (ADCC), the latter mediated largely by natural killer (NK) cells, which are depleted in heavily pre-treated patients. Adoptive transfer of NK cells from a healthy allogeneic source can restore and enhance 3F8 effects in these poor-risk patients. Natural killer (NK) cells are lymphocytes that have the capacity for antitumor activity via multiple pathways, including ADCC through engagement of the CD16 Fc receptor. The capacity of an NK cell for cytotoxic response, however, is dictated by its cell surface receptors, specifically the inhibitory killer-Ig like receptors (KIR) that are specific for self-MHC class I molecules. To maximize the effects of adoptively transferred NK cells, one should select donors from whom there is the greatest likelihood of engendering NK alloreactivity and of achieving highest functional NK response. This can be accomplished by selecting donors based on the HLA and KIR genotypes of the donor and the patient. This proposal presents a novel immunotherapeutic approach for the treatment of high-risk neuroblastoma. The combination of 3F8 monoclonal antibody with adoptively transferred NK cells from appropriately selected donors should increase ADCC, antibody efficacy, and tumor eradication. The availability of a highly effective humanized 3F8 antibody may increase tumor response while avoiding the development of neutralizing antibodies that are so common to murine analogs. Because these immunologic approaches have never been studied in combination for a solid tumor or in the pediatric population, a phase I study to examine the safety of humanized 3F8 combined with escalating doses of NK cells is necessary and represents the first aim of the proposal. Potential toxicities of NK and 3F8 therapies, a toxicity monitoring plan, dose escalation plan, the expected outcome and stopping rules are presented. The second aim seeks to correlate donor-recipient KIR/HLA immunogenetics and donor FcR polymorphism with (1) NK activation and cytotoxic function against NB target cells in vitro, and (2) with patient response to treatment. Demonstration of safety and feasibility of combining adoptively transferred NK cells with monoclonal antibody will not only result in a more potent therapeutic approach for poor- prognosis NB patients, but will be broadly applicable to other cancer populations such as lymphoma and breast carcinoma currently treated with monoclonal antibodies.